Many patients with locally advanced head and neck cancer (HNC) will be treated with curative intent chemoradiotherapy (CRT). This approach has been validated in several randomized trials and is an accepted standard of care. CRT has associated toxicity including almost universal mucositis in the radiation field. The severity of toxicity, however, varies widely with some patients experiencing only mild symptoms while others require narcotic analgesia, enteral nutrition, and interruption of therapy. Radiotherapy volumes will be predictive of the location of adverse effects but not the degree. At this time, there are no modalities that allow accurate prediction of CRT induced toxicity in HNC patients. Tissue injury induced by CRT results in regional temperature variations that can be detected by infrared (IR) imaging. Argonne National Laboratory has developed a compact computer-piloted high sensitivity IR imager with special signal detection algorithms that would allow detection of small temperature variations with little background noise. We hypothesize that alterations in thermal intensity early in therapy will be predictive of patients who will develop severe mucocutaneous toxicity. Detection of these early changes by IR imaging would identify a group of patients that will require more intensive supportive care and who may be candidates for de-intensification of therapy to avoid serious toxicity. The majority of patients with head and neck cancer will be treated with combination chemotherapy and radiotherapy and will suffer acute toxicity in the form of oral mucositis. However, there is no way to predict which patients will develop severe mucositis during therapy that often requires treatment interruption, dose reduction, narcotic analgesia, and tube feeding. We hypothesize that thermal imaging will confer the ability to predict which patients are at risk for this complication and would subsequently allow for early intervention and prevention. Such a tool would have wide applicability in this patient population and would have tremendous impact on their treatment. [unreadable] [unreadable] [unreadable]